Who am I?

• Education
  • B.Sc. Bioarchaeology (University of York, UK)
  • M.Sc. Naturwissenschaftliches Archäologie (University of Tübingen, DE)
  • Ph.D. Archaeogenetics (MPI-SHH / MPI-EVA, DE)
• Experience
  • Number of genetics classes taken: 0
  • Number of bioinformatics classes taken: 0

Today we will

1. Describe basics of DNA
2. Introduce what DNA sequencing is
3. Explain how Illumina NGS sequencing data is generated
4. How to evaluating NGS data [Practical]

What is DNA?

Deoxyribonucleic acid (/diːˈɒksɪˌraɪboʊnjuːˌkliːɪk, -ˌkleɪ-/ (DNA) is a molecule composed of two polynucleotide chains that coil around each other to form a double helix carrying genetic instructions for the development, functioning, growth and reproduction of all known organisms and many viruses. - Wikipedia

What is DNA?

What is DNA?

The rules

• Four nucleotides
  • Pyrimidines: Cytosine, Thymine
  • Purines: Guanine Adenine &
• Base pairing: one pyrimidine with one purine
  • C with G (think: CGI)
  • A with T (think: AT-AT walker)
• Complementary
  • C on one strand, G on the other (or v.v.)
  • A on one strand, T on the other (or v.v.)

The rules

• Make copy of a DNA strand with a polymerase
  • Unwind the DNA
  • Separate the strands
  • Make new strand: find a C, get new G (etc)

How do we get DNA?

What is Sequencing?

Converting the chemical nucleotides of a DNA molecule

to

ACTG on your computer screen

Historically

• Sanger sequencing

• Separate strands, add primer (starting point)
• Add mix of nucleotides, some with special ‘terminators’
• Pass through size-filtering, read order of terminators

Pros and cons of Sanger Sequencing

• Pros
  • Very precise (few errors, still the ‘gold standard’)
  • Sequence long DNA molecules
• Cons
  • Resource heavy, requiring lot of input DNA
  • Slow: one. fragment. at. a. time.

What is NGS?

• “Next Generation Sequencing”
  • Sequence millions and even billions of DNA reads at once!
  • via MASSIVE multiplexing!
  • Sequence lots of samples at once!
  • Fast and cheap!

Not really ‘next’ anymore, consider it more ‘second’ generation (see: Nanopore)

What is NGS?

(Others: Roche 454, PacBio, IonTorrent etc.)

How does it work?

• Basically same concept, but:
  • no size separation
  • with pretty pictures!

i.e. to a strand, attach a complementary fluorophore-modified nucleotide, (normally) one colour per base

A

G

T

C

Fire mah lazer, and take a picture! Rinse and repeat!

How does it work?

via Gfycat

Where does this happen?

On a ‘flow cell’

Where does this happen?

But how do you get your DNA to attach to the lawn

(and not get lost)?

• Convert it to library:
  • Add adapters: bind to the ‘lawn’ of the flow cell
  • Add indexes: sample-specific barcode
  • Add priming sites: where enzymes start copying DNA

AATGATACGGCGACCACCACaccgacaaCCCTACACGACGCTCTTCCGATCTXXXXXXAGCACACGTCTGAACTCCAGTCACgacactaCCGTCTTCTGCTTG ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| TTACTATGCCGCTGGTGGTGtggctgttGGGATGTGCTGCGAGAAGGCTAGAXXXXXXTCGTGTGCAGACTTGAGGTCAGTGctgtgatGGCAGAAGACGAAC

[Adapter & Index Primer] [Index] [Target primer] [Target] [Target primer] [Index] [Adapter & Index Primer]

Sequencing-by-synthesis

Once bound, florescence of one molecule not enough…

Make lots of copies, a.k.a. clustering! One cluster == many copies of one DNA molecule

Sequencing-by-synthesis

1. Add florescent nucleotides (complementary will bind)
2. Fire laser & take photo
3. Wash away unbound nucleotides
4. Remove fluorophore
5. Back to 1 ⤴️

What does this look like?

Improving quality

• Over time, imaging reagents get ‘tired’ and more errors occur

  • Polymerases aren’t perfect
  • Bases sometimes don’t bind, or multiple == clusters ‘desynced’
  • With each ‘photo’, machine calculates probability it got the ‘right’ nucleotide
  • Each photo of each cluster gets a ‘base-quality’ score

• What if molecule is longer than cycles of imaging?

• Improvement: paired-end sequencing

  • Get order of nucleotides by sequencing from one end
  • Get reverse order of nucleotides, by sequencing from the other end

Paired end sequencing

© 2021 Illumina, Inc. All rights reserved. Used here for training purposes only.

Photos to DNA string

• Special software (e.g. bcl2fastq):

• For each location on the flow cell (cluster):

  • Record the sequence of bases (from colours)
  • Calculates a probability the ‘base call’ is correct i.e. blurry or weak image?
  • Note the index in the sequence (sample-specific barcode)

• Group each recorded sequence or ‘reads’ with those with the same index

  • a.k.a. demultiplexing

FASTQ File

FASTQ format is a text-based format for storing both a biological sequence (usually nucleotide sequence) and its corresponding quality scores. Both the sequence letter and quality score are each encoded with a single ASCII character for brevity. - Wikipedia

FASTQ File

Example

@SRR001666.1 071112_SLXA-EAS1_s_7:5:1:817:345 length=72                      # Read ID
GGGTGATGGCCGCTGCCGATGGCGTCAAATCCCACCAAGTTACCCTTAACAACTTAAGGGTTTTCAAATAGA     # DNA sequencing
+                                                                            # Separator
IIIIIIIIIIIIIIIIIIIIIIIIIIIIII9IG9ICIIIIIIIIIIIIIIIIIIIIDIIIIIII>IIIIII/     # Quality line
@SRR001666.2 071112_SLXA-EAS1_s_7:5:1:801:338 length=72
GTTCAGGGATACGACGTTTGTATTTTAAGAATCTGAAGCAGAAGTCGATGATAATACGCGTCGTTTTATCAT
+
IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII6IBIIIIIIIIIIIIIIIIIIIIIIIGII>IIIII-I)8I

@read id

Quality score

!"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHI
0.2......................26...31........41          

Cons of NGS sequencing

• less accurate (laser/photo can get wrong)
• chemistry limits (DNA strands gets old through heat cycling for denaturing; dirty environment from suboptiomal wash steps etc.) mean short reads (compensated by volume)

Things to remember

• Indexs
• Adapters
• Cycle-quality decay
• paired-ends!